Abstract
The present meta-analysis was based on the available studies to determine the potential role of the initial and regional cerebral oxygen saturation (rSO2) in monitoring the efficiency of cardiopulmonary resuscitation (CPR) and predicting the return of spontaneous circulation (ROSC). Three electronic databases of PubMed, Embase, and the Cochrane Library were searched to identify the studies that investigated the role of rSO2 on ROSC in CA patients throughout May 2018. The weighted mean difference (WMD) with 95% confidence interval (CI) was calculated to estimate the pooled effect using a random-effects model. Sensitivity, subgroup analyses, and publication bias were conducted. A total of 13 studies involving 678 CA patients (300 in-hospital (IH) patients, and 378 out-hospital (OH) patients) were included. The summary WMD suggested that ROSC patients were associated with higher initial rSO2 (WMD: 10.10%; 95% CI: 5.66-14.55; P<0.001) and mean rSO2 (WMD: 14.16%; 95% CI: 10.51-17.81; P<0.001) levels during CA and ROSC as compared to the non-ROSC. The results of meta-regression suggested that the male percentage and the location of cardiac arrest might bias the initial or mean rSO2 and the incidence of ROSC. These significant differences were observed in nearly all subsets. The findings of this study suggested that high initial or mean rSO2 levels were both associated with an increased incidence of ROSC in CA patients undergoing CPR. These correlations might be affected by the percentage of males or the location of cardiac arrest, thereby necessitating further large-scale studies to substantiate whether these correlations differ according to gender and the location of cardiac arrest.